Abstract

The transition to electric vehicles (EVs) is gaining momentum around the world and government initiatives to accelerate this transition range from major tax exemptions, lower insurance payments to convenient parking incentives at shopping malls. The major drivers for this acceleration are the rising awareness by the public for maintaining a clean environment, reducing pollutant emissions, breaking dependencies on oil, as well as tapping into cleaner sources of energies. EVs acceptance however is hindered by several challenges; among them is their shorter driving range, slower charging rates, and the ubiquitous availability of charging locations, collectively contributing to higher anxieties for EVs drivers. Governments of developed countries as well as major car manufacturers are taking solid steps to address these challenges and set ambitious goals to make EVs the major transportation mode within few years. Consequently, a significant number of EVs is going to connect to the existing smart grid and hence, the load pattern is expecting a paradigm shift. This immense load will challenge the generation, transmission and distribution sector of the grid along with being a potential cyber-physical attack platform. To attain a graceful EV penetration for curtailing GHG emission, along with the socioeconomic initiatives, an extensive research is required, especially to mitigate the range anxiety and ameliorate the load congestion on the grid. As a consequence, to reduce the range anxiety, we present a two-stage solution to provision and dimension a DC fast charging station (CS) network for the anticipated energy demand and that minimizes the deployment cost while ensuring a certain quality of experience for charging e.g., acceptable waiting times and shorter travel distances to charge. This solution also maintains the voltage stability by considering the distribution grid capacity, determining transformers’ rating to support peak demand of EV charging and adding a minimum number of voltage regulators based on the impact over the power distribution network. We propose, evaluate and compare two CS network expansion models to determine a cost-effective and adaptive CSs provisioning solution that can efficiently expand the CS network to accommodate future EV charging and conventional load demands. Though an adequate fast charging network may assist to reduce the range anxiety and propel the EV market, catering this large number of EVs using fuel based conventional grid actually shifts the carbon footprint from the transportation sector to the power generation sector. As a consequence, green energy needs to be promoted for EV charging. However, the intermittent behavior of renewable energy (RE) generation challenges to maintain a RE based stand alone CS. In order to address this issue, we consider a photovoltaic(PV) powered station equipped with an energy storage system (ESS), which is assumed to be capable of assigning variable charging rates to different EVs to fulfill their demands inside their declared deadlines at minimum price. To ensure fairness, a charging rate dependent pricing mechanism is proposed to assure a higher price for enjoying a higher charging rate. The PV generation profile and future load request are forecasted at each time slot, to handle the respective uncertainties. Whatever, the energy source is green or not of a CS, a static CS cannot offer the flexibility to charge an EV at any place at any time especially for an emergency case. Fortunately, the bidirectional energy transferring capability between vehicles (i.e., vehicle to vehicle (V2V)) might be a solution to charge an EV at any place and at any time without leaning on a stationary CS. Hence, we assume a market where charging providers each has a number of charging trucks equipped with a larger battery and a fast charger to charge a number of EVs at some particular parking lots. We formulate an integer linear program (ILP) to maximize the number of served EVs by determining the optimal trajectory and schedule of each truck. Owing to its complexity, we implement Dantzig-Wolfe decomposition approach to solve this. However, to build a prolific EV charging ecosystem, all its entities (e.g., EVs, CSs and grid) have to be connected through a communication link and that unveils a new cyber physical attack surface. As a consequence, we exploit the abundance of Electric Vehicles (EVs) to target the stability of the power grid by presenting a realistic coordinated switching attack that initiates inter-area oscillations between different areas of the power grid and assess the dire consequences over the power system. Finally, a back propagation neural network (BPNN) technique is used in a proposed framework to detect such switching attacks before being executed.